Catching internet trolls

We can never be sure that the opinion we are endeavoring to stifle is a false opinion; and if we were sure, stifling it would be an evil still.  John Stuart Mill, On Liberty, 1859

In the memokraat blog, the Short Guide to Catching Trolls (Lühike trolliküti käsiraamat ehk ettepanekud online diskussioonitehnoloogia arendamiseks Eestis) was posted discussing the issue of trolling at the reader comment sections of Estonian online news media sites. I briefly participated in the debate, but now will provide here an overview of some issues and questions that I have been thinking about since then.

1. Defining the troll. The Guide defines troll as someone who posts insulting comments in an online environment with the purpose of provoking others and disturb the discussion. What constitutes an insult, provocation or disturbance is subjective to a certain level and depends on the cultural and social background of a person. My comments for the Guide as posted in the comment thread could be considered trolling as at least some people thought that part of my arguments were arrogant (when I suggested that people who are so thin-skinned that they could not tolerate hostile commentary should not read it), the style I write and express my opinions is usually a little provocational and some people might think that it disturbed the discussion. Thus in a wide reading of the definition I am a troll and so is nearly anyone who does not conform to a certain subjective criteria defined by the owner of the site. The banning of such trolls as myself results in discussions that are perhaps non-insulting and uncontroversial, but at the same time also rather valueless in terms of expanding the scope of discussion and bringing in alternative views.

2. Defining the problems? The trollhunters claim that the problems are caused by the technology used for commenting at Estonian online news sites. They claim the technology used is to blame for the exclusion of certain other people (who in a bizarre twist are claimed to be unable to freely use their freedom of speech because they are afraid of others also using it but critically), that this leads to radicalisation of public opinion and intolerance and distorts public opinion. All these claims lack evidence, studies or any research: we are expected to accept these premises as self-evident, when they are actually not.

The biggest unsupported assumption the authors of the Guide make* is that the root of all evil is anonymous commenting, which may or may not be the case.

3. Freedom of speech in an online setting. It is true that freedom of speech is not absolute and the exercise of it also requires responsibility for one’s opinions. However, for acts that bring criminal liability there already is a possibility to identify almost anyone online so the anonymity is only superficial. Thus it is possible at least in theory to make people accountable for their words also now.

50 years ago internet did not exist and therefore the human rights standards we have need to be adopted to the internet era. Certain principles remain the same, but the internet might change the content of certain rights subtantially, including freedom of speech. Offline analogies do not always work in the global unregulated internet with low barriers to entry.

4. Self-regulation is not always best when dealing with human rights issues. The trollhunters state that the least they want is state intervention or regulation. They propose a system of self-regulation, whereby the parties who control the commentary space make an agreement which is then adopted and implemented. In my mind it is dangerous when private entities make deals that involve limits to freedom of speech, because questions of accountability and transparency rise. I also believe that this might be even worse than state regulation, because the state is much more bound by international human rights obligations. So in cases that involve human rights I think it is preferential to have state regulation, rather than allow for private parties who control substantial public discussion space to make their own agreements to limit certain aspects of online activities. In many many spheres self-regulation is possible and works very well, but I am not sure if this is it.

—

In general terms there is another fundamental point why I think any regulation of online commentary space is not beneficial. Net neutrality means that content providers should not restrict specific parts of the internet depending on subjective criteria. The internet has flourished partly because of the freedom it provides and the abscence of walled gardens. What the trollhunters want to create is a walled garden, admittedly with very low walls, but still access to commenting would become more limited.

I also disagree that this discussion should be framed in terms of media freedom. Online commentaries do not constitute media or journalism in my opinion, it is a separate issue and therefore references to media and press freedom indexes do not really matter. The public debate in the matter has also included mostly people from the media sphere and not other areas who have framed the debate in their own terms.

Short rebuttal of the trollhunters critique of anonymity online:

1. The whistle-blower effect. The trollhunters claim that an anonymous tip option might be sufficient cannot be accepted because it is media-centric and would require an interested journalist to pursue the topic and lack of self-censorship in the media. Anonymous hints cannot replace anonymous commentary, they are different things. The trollhunters agree that certain levels of anonymity might be necessary for informing the public, I say it is essential.

2. Balance between insulting and constructive comments. This depends on the topic, but in most cases I would say based on my experience that insulting comments do not prevail over more insightful ones. Again, there are no studies made or statistical data available other than Delfi claiming that insulting comments are only a fraction of the total body of comments. Even if the majority of comments would bring nothing to the debate, is it worth not having the few that do?

3. Censorship. The trollhunters here refer to the right of newspaper editors to choose what to publish. The commentary space is not in my understanding part of the newspaper and can function without it. The fact that newspapers have always edited their stories have been due to physical lack of room in the paper and the need to provide a concentrated overview. The internet does not have these physical limits and there is nothing lost with adding to concentrated overviews and officially sanctioned opinions other stuff as well. The claims of this leading to mob censorship are speculation.

4. Impact of insulting anonymous comments. It is true that some people are more sensitive to criticism and insults than others, but again, there is no need to censor everyone because of this. Some views are controversial and people get offended. If I want to claim that there is no god, then this is offends people and I will be branded a troll. I believe that online commentary space must allow for expression of those less conformist views, which are not published by newspapers. I disagree that it is right to take away the freedom of expression from one group for the benefit of another.

5. Strength or importance of message. See above. I still refuse to accept the approach that it is somehow justified to prefer one group of people to another in terms of who may or may not express their views.

6. Vox populi, vox dei. I agree with the trollhunters that there needs to be no correlation between general public opinion and views expressed in comments, but I think it also depends on the topic. Likewise, I have seen no statistics which confirm that people believe that views expressed in online commentaries represent the general views in the society.

7. Video game violence argument. The idea is that people can insult others virtually so they will not do it offline. I think there is no correlation here, but no studies have been provided saying one thing or the other. The anger people have is in my own opinion a consequence, and not a cause of the processes in the society.

8. Moderated comments loses valuable discussion. This is subjective. If in some specialist finance related forum it works and likeminded people can express themselves better and feel good about it then fine. I do not think that discussion space for general public should work the same way (the danger to stifling of dissent and danger of conformism).

9. Too many comments to moderate. I think the notice and take-down system works pretty well.

10. Topic already discussed. Nothing to say here. Everything should be open for discussion.

11. Notice and take-down. Probably the system could be improved, I think it is disproportional to ban all anonymous comments due to imperfections of the notice and take-down system.

12. Turn to police. In case there is no real threat police should not be involved. If a person says to another that go jump off a cliff then this obviously is not a real threat. Again, in those cases where there is real threat police should be involved. It does not mean that all anonymous comments need to be banned because police does not do their work.

13. Economic factor. I am not sure if the commenting option is ecnomically beneficial or not, but of course that should not be the prevailing argument for or against limiting free speech online.

14. Freedom of speech is why comments are kept. Not a convincing argument, I agree. But I do think that once they have been introduced, stopping the ability to comment wold be problematic as it has almost become a service of general public interest.

15. Historical perspectives. There has been no time like this and therefore parallels with the past might or might not provide insight into dealing with the issue at hand.

* As pointed out by Daniel, I have put words in the mouths of the trollhunters that they think that anonymous commenting is the root of all evil. The trollhunter guide does not state this and it was an exaggeration on my part. However, I still think that the general tone of the Guide seems to connect anonymous expression with the existence of insulting and derogatory comments. I guess it would be fair to say that the authors consider it as a not an insignificant part of all evil 🙂